Influence of TV Media Content on Display Lifetime and Image Sticking Measurement Techniques
Andrew W. Johnson, Marie O’Regan, Johann Trujillo
DuPont Displays, Santa Barbara CA安全教育小班
Andrew.Johnson-1@
Abstract
Both lifetime and image sticking measurement techniques are well established but are somewhat uncharacteristic of actual TV usage. The purpo of this paper is to quantify TV usage under various media categories and propo modifications to the standard experimental techniques to better reprent the usage demands from TV media.
Author Keywords
Lifetime Measurements; Image Sticking Measurements
1.Introduction and Objective
As development of AMOLED technology progress from smart phone size displays towards the TV market, lifetime requirements become more demanding. Luminance decay during operating life is the standard degradation mechanism of OLEDs, and thus a potential barrier to commercialization. Therefore it is important to establish the lifetime and resistance to image sticking requirements for high product quality. The standards developed must be carefully considered becau, while non-rigorous standards will cau unhappy customers and market failure after commercialization, overly rigorous requirements could cau excessive development and misd market opportunities. Prently, both lifetime and image sticking measurement techniques are fairly well established, but not reprentative of actual TV usage. For example, lifetime testing primarily consists of measuring luminance decay at max brightness (RGB grayscale=255) for 100% of the display1. Image sticking is classically measured with a white square operating at 100% of max brightness at the display’s center with a black background (RGB grayscale=0)1. Neither condition reprents any normal TV content.
The purpo of this paper is to quantify TV usage under various media categories and propo modifications to the standard experimental techniques to better reprent the usage demands from TV media.
三角形的边长
2. Methods
To characterize TV usage, the following media categories were analyzed for sub-pixel intensity, color, and overall image uniformity: TV shows, news, sports, video games, and movies. All of the categories contained a wide range of genres. For movies, this encompasd animation with bright colors, documentaries with lots of text, outdoor movies with bright sunlight, and dark movies. The analysis of news stations included all of the major networks and two cable stations. Sports media encompasd both indoor and outdoor venues. Video games entailed role playing, 1st person shooting, and sports games. Finally, a variety of TV genre including sitcom, reality, drama were studied.
From each source of media, a large sample t of images were extracted and analyzed. To reduce the large quantities of data, the images were reduced to a 320x180 pixel resolution before final pixel-by-pixel analysis was done. The color and intensity of each pixel in each frame was measured using the sRGB color space and 8 bit resolution. A 2D array of TV usage was generated from the large arrays of intensity and color by averaging all the frames together.
3. Results运动会音乐
(a) Pixel Intensity: Life testing a display at max brightness is uful in understanding the upper limits of its capability. However, it is more practical to know the actual extent of the product’s commercial life by testing under more reprentative conditions. Typical pixel operating conditions are depicted in Table 1 in terms 255 RGB grayscale.
Table 1: Mean RGB grayscale results for different media
订餐genres
Media Genre Red Results
(max=255)
Green
Results
(max= 255)
Blue Results
(max= 255) TV Shows 86 77 70 Sports 105 107 91 News 80 78 82 Video Game 79 77 65 Movies 78 75 67
Upper
Grayscale
Setting 105 107 91
The intensity values for movies are consistent with other measurements, reinforcing the industry practice of using 30%2 values, or 75/255, to project actual TV lifetime. The sports category required the brightest picture of all the media forms tested and the video games were the dimmest. Using the highest values for each color gives a maximum grayscale for red, green, blue as 105, 107, 91 respectively. The range of intensities between media content should be considered when estimating a TV’s lifetime. Within this range, a more conrvative approach would u the highest values of around 107 for all three colors. A more practical value that reflects a wider range of viewing habits would be the median for RGB which is approximately a grayscale tting of 75.
(b) Image Sticking:Image sticking, or a residual image that can be en when the original image is n
o longer being displayed, can come from a variety of content sources. Video games are well known for leaving residual images of game status indicators. For news stations, the main culprit is a logo or band along the bottom of the screen. Sports logos and scoreboards may also leave image artifacts.
6.1 / A. Johnson
SID 2012 DIGEST • 43
ISSN 0097-966X/12/4301-0043-$1.00 © 2012 SID
(a) Top-down view of soccer game’s green channel
(b) 3D view of soccer game’s green channel
Figure 1. 3D graphs (column, row, grayscale intensity) created from merged images of a soccer game’s green channel.
助动词的用法口诀
Figure 1 shows the green channel composite images from a soccer game. The scoreboard can be clearly en at the left graph’s upper left corner in the figure. This graph shows a top-down view of a 3D chart (axes: column, row, grayscale intensity). The graph on the right shows the difference in magnitude between the general content and the scoreboard. The grayscale level difference between the background content and the bright region for all three colors is approximately 64/255 vs 200/255 respectively.
Table 2 shows a summary of the various media content and the typical relative differences between the image sticking sources with respect to the background content.
Table 2: For different media genre, RGB average grayscale level for the image sticking sources and
background content.
发征Media Genre
Red Sticking Source 255 Grayscale
Level (Source vs Background)
Green Sticking
Source 255 Grayscale
Level (Source vs Background)
Blue Sticking Source 255 Grayscale
Level (Source vs Background)
TV Shows 133 vs 86 124 vs 77 119 vs 70 Sports 183 vs 105 181 vs 107 180 vs 91 News 170 vs 80 163 vs 78 170 vs 82 Video Game 176 vs 79 176 vs 77 174 vs 65 Movies 78 vs 78 75 vs 75 67 vs 67 Max Ratio 176 vs 79
176 vs 77
174 vs 65
News broadcasts and video games have large sources of image sticking while movies contain no sources. The relative difference between bright areas and the background for the worst ca scenarios is approximately a 175 to 74 difference. This is significantly less stringent then the recommended IEC 255 to 0
pattern. This discrepancy between test condition and actual usage pattern will lead to a poor prediction of display performance from test results.
(c) Color Usage: Understanding what colors are typically
displayed is important for both life time and image sticking. Since red, green, and blue OLED sub-pixels typically have different decay rates, the most common color en by a ur should influence how a display is tested. For example, if white is ud a significant amount of the time, then lifetime and image sticking should both be tested and reported under white conditions. As a result, both luminance decay and color change will be critical data to measure. Converly if the display is to be ud to show pure red, green or blue, then individual color testing becomes more important.
As Table 3 indicates, the dominant color displayed under all content genres is either white or some shade of it: yellow or bluish white as defined by the 1931 CIE Chromaticity Diagram. Pure red, green,
and blue are either ud in small quantities or for short periods of time.
Table 3: Color usage by Genre
Media Genre
White % of total color
% of total color that is some variation of white
% of total color that is pure red, green, & blue
TV Shows 63% 88% 0.2%, 0.0%, 0.2% Sports
55% 77% 0.3%, 0.0%, 0.1% News 42% 72% 1.2%, 0.3%, 1.2% Video Game 69% 89% 0.1%, 0.1%, 0.0% Movies
华山之巅
45%
80%
0.7%, 1.0%, 0.7%
From the data, it makes n to test both lifetime and image sticking using white patterns rather than with individual colors.
6.1 / A. Johnson
44 • SID 2012 DIGEST
4. Impact
(a) Lifetime Measurement: Most media categories operate at a 75/255 grayscale range for all three colors. Under the more luminance inten sports media, this level was measured to be higher and around 107/255. Life estimations may need to be modified depending on the expected u of luminance inten applications. Accelerated lifetime studies done at max brightness can then be related back to the non-accelerated conditions, which are significantly lower. Since a majority of media content is some variation of white, test measurements should be reported under white conditions, paying clo attention to both luminance and color changes.
(b) Image Sticking Measurement:While the IEC recommended pattern and measurement technique is good at capturing image sticking in TV media, the practice of comparing a 100% brightness (255/2
55) to a black background (0/255) is uncharacteristic of TV media and therefore an overly stringent test. The more inten image sticking possibilities that exist from news, video games, and sports media, would be better reprented by a 175/255 square and 74/255 background. We suggest that the background and square should be operated at both this level and at the specified white point.
5. Conclusions
Extensive examination of typical TV content has shown that display aging and image sticking test methods should be reconsidered. The typical estimation that a display will operate at 75/255 grayscale may be low in certain applications. Image sticking may not be as vere a problem as once thought since typical grayscale level differences are in the range 175/255 vs 75/255 as oppod to 255/255 vs 0/255.
6.References
[1]INTERNATIONAL ELECTROTECHNICAL
COMMISSION: ORGANIC LIGHT EMITTING DIODE
(OLED) DISPLAYS- Part 5-3: Measuring Methods of Image Sticking and Lifetime.
水芥菜
[2]Young, Barry, “OLEDs- Promis, Myths, and TVs,”
Information Display, 9/09, 14-17 (2009)
6.1 / A. Johnson
SID 2012 DIGEST • 45
